Achieving Reliable Access to Water in Rural Communities
A Study for the Water Compass
Authors: Marije van den Broek and Dr. Julia Brown [email protected] [email protected]
Figure 1: Children collecting water from a valley dam
1
Contents page
Executive Summary…………………………………………………………....3
1. Introduction and Rationale………………………………………………...6
2. Background…………………………………………………………………..8 2.1. Country overview of rural water provision and policy in Uganda…………….. 8 2.2. Gomba District……………………………………………………………………... 9
3. Achieving Reliable Access to Water in Rural Communities………..11 3.1. The Community Based Management (CBM) model…………………………..12 3.2 The case for investing in the sustainability of rural water services………….. 13 3.3 What are the recurrent costs of handpumps?...... 14 3.4 Rural Water Management Models………………………………………………. 15
4. Study Methods …………………………………………………………….. 17
5. Findings……………………………………………………………………....23 5.1 Assessment of current water arrangements and management practices ……23 5.2 Selection of the water management model……………………………………..28 5.3 Visiting Demonstration Sites……………………………………………………....34
6. Recommendations and Conclusions……………………………………37
References………………………………………………………………………40
Appendix 1 Water Management Models……………………………………42
2
Executive summary
The Water Compass is a Water, Sanitation, and Hygiene (WASH) organisation operational in Gomba District in Uganda. The ambition of the organisation is to improve access to water in rural sub-Saharan regions with a special emphasis on post-construction sustainability of boreholes equipped with a handpump. The focus of the Water Compass on post-construction sustainability prior to hardware construction represents a unique and forward-thinking approach that differentiates the organisation from other NGOs in the rural water sector.
The poor functionality performance of constructed sources is a concern for the sector in general and the focus of academic studies by van den Broek and Brown. It is commonly acknowledged that across SSA at any one time a third of handpumps are non-functional (Baumann, 2006; Oxford/RFL, 2014; RWSN,2010). There are different schools of thought to explain the woeful sustainability record of constructed water sources – from poor siting and construction techniques, to lack of spare parts, to the failings of the post-construction management model itself (most commonly Community Based Management - CBM) and lack of collected funds to pay for maintenance. Van den Broek and Brown’s work strongly suggests that the last mentioned is the dominant explanation behind the poor performance of the rural water sector in the areas they have studied in Uganda.
To facilitate long-term and reliable access to rural water facilities, the Water Compass commissioned the University of Portsmouth to provide a situation assessment of current water access and practices, and simultaneously to undertake a study on the effectiveness of different rural water management models for facilitating sustainable operation and maintenance practices of handpumps in the local context of Gomba District.
Fieldwork in 12 villages in the two sub-counties of the Water Compass’s Gomba operational area, across a two-week period at the end of the dry season in February 2017, provided valuable information about the current status of safe water provision. Safe water provision is a priority in Maddu and Kabulasoke sub-counties – water is currently either being collected from untreated surface sources (valley dams and tanks), shared with livestock at a distance from dwellings, or for many it is a commodity: utilising water vendors charging considerable sums for water without assurance of its origin is common practice. There are some handpumps attached to boreholes but coverage is still low. Overall, there
3 is a clear need for more safe water sources in the area – the topography, however, with a low water table means that siting of sources may mean compromises are made – fewer more reliable sources at a distance from dwellings.
The participatory nature of the research involved gathering community views on how best to manage constructed water sources for long term sustainability, and opinions were sought on the viability of 5 post-construction management systems currently used or piloted across Uganda that van den Broek and Brown identified as potential management models. Local feasibility, acceptance and support for post-construction maintenance is crucial – hence the need to incorporate local views and suggested modifications. The five management models investigated were:
1. The current government policy of Community Based Management (CBM);
2. Modification of CBM with the addition of a Village Savings And Loan Association (VSLA) system;
3. CBM-lite which replaces the voluntary management system with a paid water operator who stores collected funds in a micro-finance account that offers an insurance-style product to expedite repairs;
4. Mobile Money. In this model a community member collects funds from water users and sends them to a mobile money account to be stored for maintenance;
5. WaterTime is a pre-paid mobile-enabled technological innovation fitted to a handpump and management system that uses familiarity with purchasing mobile phone credit, ‘airtime’, to depersonalise payment for accessing water. Collected funds are transferred to a bank account by mobile money.
In all 12 villages participants voted overwhelmingly for the fifth model – WaterTime. The strength of support for WaterTime was a surprising outcome of the study. Community members envisaged issues over collecting funds via the CBM model or VSLA – the potential for misuse of funds or confrontations over money led people to support a model that depersonalises payment and management by favouring individual over collective responsibility. WaterTime also provides a potential business opportunity in rural areas. In addition, the WaterTime model attracted support from the District Water Officer and Assistant Water Officer and a representative of the Gomba District Handpump Mechanics Association.
4
The new technology is being piloted by two NGOs at two locations near Fort Portal in western Uganda – there is proof of technological concept, but there are aspects of the management system that could be strengthened. Familiarity with paying for mobile phone credit – known regionally as “airtime” - led to the coining of WaterTime – water users purchase water credits, like mobile phone credit, from village kiosks.
There are cost implications for the new technology, which we would recommend is installed during construction rather than at a later date (although the technology is designed to be retro-fitted). Introducing a stricter payment model may be more challenging if users are accustomed to circumventing payment and free-riding. We recommend piloting the new technology in carefully selected locations so that lessons can be learnt. The transparent and watertight management of the collected funds is vital for the long-term future of the WaterTime Model. Water Compass will need to manage the system before transfer to a trusted private operator – while still providing oversight.
The philosophy of the Water Compass to embrace careful testing of innovations is what the traditionally risk averse rural water sector needs – to learn from doing and to share findings (positive or negative) is vital. Our previous studies indicate that there is not a one- size-fits-all model for rural water management – that local conditions, such as the availability of alternative sources, may mean that a different approach may be suitable in different contexts.
5
1. Introduction
The Water Compass is a new WASH NGO, registered in the USA and operating in Uganda, with a philosophy of planning for post-construction sustainability prior to commencing operations. Gomba District was selected as their initial area of operation because nationally it has a low safe water coverage (56.1%) and has not attracted many NGOs to support the operations of the government in providing safe water sources and improvements to sanitation. The Water Compass is working closely with the District Water Office and has the support of the Rural Water Supply Department in the Directorate of Water Development, Ministry of Water and Environment.
Marije van den Broek, prior to commencing PhD studies at the University of Portsmouth with Dr. Julia Brown, was a Programme Manager for a Water and Sanitation and Hygiene (WASH) NGO in mid-West Uganda, with previous experience in the WASH sector in northern Uganda, and brings a wealth of experience concerning different management models. Van den Broek and Brown have collaborated since 2012 on evaluating rural water management models, with a focus on the predominant model – Community Based Management - and alternatives, and have published on the topic. The Water Compass commissioned van den Broek and Brown to undertake a study that would help Water Compass in developing a post-construction management strategy for their area of operation.
This report has two main goals. First, to provide a situation assessment of current water access and rural water management practices in the two sub-counties of Gomba District where Water Compass will be operating (Maddu and Kabulosoke). This assessment of water coverage has spatial and temporal limits: 12 villages in two sub-counties (6 apiece) were studied over the course of two weeks towards the end of the dry season - February 2017. The study does, however, provide an indication of the current and potential issues, despite a small sample size of villages, because of a good level of community attendance and engagement in the research process. In addition, meetings were also held with officials from the Ministry of Water and the Environment Ministry officials responsible for operation and maintenance in the rural water sector, Gomba District Water Officer and Assistant District Water Office; political leader of Maddu sub-county (Local Councillor 3, or LC3) and sub-county chief of Maddu (government appointed technical staff) and a representative of the district Handpump Mechanics Association. Once the Gomba fieldwork was completed, a visit was also made to Fort Portal to visit two sites where two Ugandan NGOs (Joint Effort to Save the Environment – JESE – and Health Through Water
6 and Sanitation - HEWASA) who work closely together, are piloting a new technology designed by Practica Foundation and Susteq, both from the Netherlands.
The second aim was to provide recommendations on possible post-construction management strategies Water Compass could pursue for sustainable rural water management. The study provides an overview of the costs of maintaining a deep borehole which makes the need for a reliable water management model that guarantees available funds to cover maintenance and repair costs an imperative for long term sustainability.
Imposing management models in a wholly top-down fashion has been proven to be problematic – there is a need to factor in local conditions (Adams, 1992 and Chambers, 1997) - and local support is also beneficial before introducing new systems (Brown and van den Broek, 2017). We first scoped from participants their ideas for ensuring post- construction sustainability of water sources. In order to make the recommendations feasible we also presented five management models currently in operation or being piloted within Uganda – these were selected prior to commencing fieldwork, utilising recent PhD studies by van den Broek. After presenting the models, we encouraged participants to ask questions and consider how viable the models may be in their context and what modifications may be useful – we wanted any recommendation to be as user-driven as possible. The overall aim of the research was to be participatory and to generate locally supported and realistic recommendations for the Water Compass. The report also makes an assessment of the key considerations for piloting the model that received the most local support – we are terming it, as a result of fieldwork – WaterTime.
The report now proceeds with a brief overview of rural water provision and the policy framework in Uganda. We then provide key highlights of the water and sanitation situation in Gomba, drawing on official documents. This is followed by a summary of the dominant management model in Uganda – Community Based Management (CBM) - and the case for investing in the sustainability of rural water services and considering alternatives to mainstream CBM. We then provide an overview of five rural water management models currently practised and piloted in Uganda. Then we describe the participatory study and research approach undertaken. The research findings are divided into our assessment of the current water arrangements and management practices in Gomba District, and the preferred water management model selected by community members in the 12 study villages. Finally, we consider the findings and their applicability for the Water Compass, and make a series of recommendations.
7
2. Background
2.1 Country overview of rural water provision and policy in Uganda.
Uganda, in central East Africa, is still predominately rural – with 85% of the population residing in rural areas. By 2025 Uganda is predicted to experience water stress due to its rapid population growth (at 3.24% in 2014, it has one of the highest rates in the world – CIA World Fact Book1) and degradation of wetlands that are vital in regulating groundwater systems that large swathes of the population are dependent upon for safe drinking water (Wong et al 2005). Climate change is also beginning to result in changes to the traditional dry and rainy seasons with a significant impact on family activities and water availability2.
Important aspects of Uganda’s legal framework include: • The Constitution of Uganda (1995), revised 2005, which enshrined the right of all Ugandans to clean water. • Local Government Act (1997) that saw the devolution of water provision, maintenance and follow-up support responsibility to local governments. • Under the National Water Policy (1999) communities are responsible for the management of their water sources and are expected to establish a Water User Committee – a voluntary group who collect the water user fees (the users pays principle is a central component of Uganda’s water policy framework) in keeping with the Community Based Management model (Ministry of Water and Environment Uganda, 2007).
Government targets were for 77% of the rural population to have access to an improved water point by 2015, and 90% of constructed handpumps to be functional. The Uganda Ministry of Water and Environment (MWE) in 2014 reported 64% coverage and 85% functionality. However, it is argued official figures over-report functionality (Burr and Fonseca, 2013) – in Kanungu District; officially 78% were functional, whereas the study of Koestler et al. (2010) report 40%.
1 https://www.cia.gov/library/publications/the-world-factbook/ 2 http://www.gcca.eu/national-programmes/africa/gcca-uganda
8
2.2 Gomba District
Gomba District, in central Uganda, was formed from Mpigi District in 2010 and has its District headquarters in Kanoni town, 93KM from Kampala (around 2.5 hours by road). It comprises 2 counties (Gomba East and West) and 4 sub-counties – Mpenja, Kyegonza, Maddu and Kabulosoke – the latter two are the focus of this present study as they are the planned operational area of the Water Compass. The District has a population of 160,075 (2014) residing in 37 parishes and the 289 villages.
Gomba, due to its orientation, experiences both equatorial and tropical climate, as well as the regionally typical two rainy seasons (peaks of which occur March to May, and then between September to November). Vegetation patterns match prevailing climates – with dense equatorial forests in the east and north of the District, with tropical grasslands dominating the south and south-west. Zooming into our study area, subsistence farming is the predominant livelihood in western Gomba, with cattle keeping in the west and north west (notably Maddu sub-county). In terms of terrain, Gomba is described as “hilly with wide and narrow v-shaped valleys with perennial streams” (part of the River Katonga complex). The south west of the District is flatter with swampland and a relatively higher water table, and forms part of the Lake Victoria Basin3. The topography is important in this study because the water table is generally low which has water provision implications – often necessitating the construction of deep-bore holes to access low-lying ground water. The current range of water sources in the District is set out in Table 1.
According to the Water Supply Atlas, Gomba District currently has 708 domestic water points, serving a population of 133,597 people, with the majority (121,314) in rural areas. There is one piped water system serving Maddu town (undergoing maintenance during the time of fieldwork). The Water Supply Atlas also reports that 154 water points are classified as abandoned because they have been non-functional for over 5 years4.
3 Gomba District 2016-2017 Annual Work Plan: Budget for rural water and sanitation development conditional grant. 4 http://www.wateruganda.com/index.php/reports/district/95
9
Table 1: Range of water sources in Gomba 2017 (source: Gomba District Water Office)
Water Source Protected and unprotected springs Groundwater
Shallow Handdug Wells – with Groundwater handpump(where topography allows)
Boreholes – with handpump Groundwater Valley dams (collecting run-off) Untreated surface water Valley tanks (collecting run-off) Untreated surface water Ponds, rivers, steams and lakes Untreated surface water 1 Piped water scheme Production well
Access to water varies across the District – from a high of 95 % in Mpenja to 56% in Maddu- sub-county. Overall safe water coverage in Gomba is 56.1% - with some variation as detailed in Table 2 below.
Table 2: Safe water coverage in Gomba 2017 (source: Gomba District Water Office) Gomba Sub-counties Population served % safe Water Mpenja 32,114 71.1 Kabulasoke 49,991 40.5 Kyegonza 30,372 44.0 Maddu 35,115 50.0
Table 3: 2016-17 Annual work plan/budget for water supply and quality (source: Gomba District Water Office)
Category Uganda Shillings US Dollar Water supply facilities 225,000,000 $78,9475 (hardware) Rehabilitation of water 43,000,000 (of which $15,087 facilities 7,800,000 would be for water testing) Water quality surveillance 3,877,500 $1,360
The budget highlights the fact that the District does not have the funds to radically extend the supply network. Further implications of the budget are that the District does not have the funds to undertake extensive rehabilitation of sources. Given 154 sources are currently classified as non-functional this means that effectively many sources will have to wait a protracted period before the District can attend to them. The recurring costs of maintenance and rehabilitation are discussed shortly in section 3.3.
5 USD 1= 2850 Uganda Shilling (UGX)
10
While not the specific focus of this study, access to protected water/clean water needs to be provided in conjunction with enhancements to sanitation situations. Table 4 sets out the current sanitation situation in Gomba District.
Table 4: 2017 Sanitation situation (source Gomba District Water Office).
Gomba Sub-Counties % pit latrines % hand washing Mpenja 67 31 Kabulasoke 56 20 Kyegonza 53 20 Maddu 42 12
Table 5: 2016-17 Annual work plan/budget for sanitation (source District Water Office)
Uganda Shilling US Dollar Sanitation 17,856,000 $ 6,265
Given the low level of pit latrine coverage, the District budget is fairly meagre. As with the budget for water provision, this underlines that there is an important role for the Water Compass in working closely with and complementing, the activities of the District Water Office.
3. Achieving Reliable Access to Water in Rural Communities
In this section we will provide a summary of the dominant post-construction management model – Community Based Management (CBM) - introduced in Uganda in 1986 by UNICEF. The disappointing results of CBM across Sun-Saharan Africa (SSA) have led some in the rural water sector to investigate alternatives to CBM that may yield improved functionality rates. We set out the arguments to explore alternatives. We then briefly set out the costs of maintenance and rehabilitation: whichever management model is in place, it needs to be able to generate funds to cover these projected costs. Finally, we present and justify the five management models we selected to test the level of community support and overall feasibility in Gomba District.
11
3.1 The Community Based Management (CBM) model
Since the 1980s, Community Based Management has been promoted as the best option to realise sustainable access to water across Sub-Saharan Africa: it is dominant management model for rural ground water management. With CBM, communities demand a water service and contribute to its construction (usually 5-10%) and post construction, mange the source through the establishment of a voluntary Water User Committee who arrange for the collection of user fees for its upkeep and maintenance and make arrangements for the handpump mechanic (Briscoe and Ferranti, 1988). With maintenance of sources the responsibility of the users, in keeping with the user pays principle, governments and donors focused their attention on extending the supply network. However, and depressingly, the post-construction functionality rates of community managed sources are the same as they were during the acknowledged poorly performing state-led era: a third of handpumps are still non functional at any one time (Baumann, 2006; Oxford/RFL, 2014; RWSN). Our research in other districts in Uganda (Masindi and Kiryandongo) found that out of 100 sources just three had adequate funds for major repairs, and 53 had not collected funds at all: during the state-led and community-led era, maintenance has been, and continues to be, the stumbling block (van den Broek and Brown, 2015).
The CBM model has been strenuously promoted by the international community, and has indeed achieved satisfactory service levels in higher income countries (Hutchings et al., 2015), in water scarce areas (due to the lack of alternatives) and in contexts with frequent external support (Smits et al., 2013). However, the two key pillars of the CBM model – voluntary community control and user payment – in the sub-Saharan context often appear irreconcilable and a source of conflict. Here close social relations and the informal and voluntary management arrangements in the CBM model frequently prohibit the regular collection of user funds (van den Broek and Brown, 2015). A number of studies have demonstrated that low-income households are able to afford the water user fee (an average lump sum of USD 0.35 per month in Uganda), but in reality free-ride as a result of the informal management structures and collection methods (Fonseca, 2014; van den Broek and Brown, 2015). The problem of non-payment is therefore largely related to water users’ unwillingness to pay rather than their inability to pay6 (Jones, 2011; Burr and
6 There are groups for whom payment and affordability is an issue. In other study areas a list of those households exempt is kept - female-headed households, elderly and disabled (Brown and van den
12
Fonseca, 2013). Besides the irregular income for operation and maintenance, a major problem is the mismanagement of any collected water user fees by members of the Water User Committee. Experiences of misuse of perceptions of misuse are then used to justify non-payment. Ultimately, the lack of funds often cause long delays in repairs and tend to put community members at considerable risk as, in the absence of nearby protected water points, people may resort to open and contaminated water sources (van den Broek and Brown, 2015).
3.2 The case for investing in the sustainability of rural water services
Reliable access to water in rural Sub-Saharan Africa (SSA) represents a long-standing challenge. While the majority of the people in rural SSA are dependent on handpumps for accessing their drinking water, estimates indicate that at least a third of these water facilities are non-functional (RWSN, 2010; Oxford/RFL, 2014). The high breakdown rate of handpumps is causing serious health risks to communities and involves enormous financial losses to governments and donors, estimated at USD 1.2-1.5 billion over the last two decades (Baumann, 2006).
The handpump was introduced in the 1980s due to its low-cost technology and simple maintenance requirements. Despite the low-cost technology compared to other water service levels such as piped water schemes, handpumps require frequent repair and (small) regular financial injections to ensure the upkeep of the system.
As discussed, since the 1980s, rural communities in developing countries are charged with operating, paying and voluntarily maintaining communal water sources. However, a growing body of research is demonstrating that community management is a prime contributor to the disappointing functionality rate of rural handpumps. Communities tend to have a chronic lack of funds to pay for maintenance and repair and informal community management practices are often a source of internal conflict. The need to pay water user fees often creates tensions, and the insurmountable challenges facing the Water User Committees in enforcing rules results in a general aversion to volunteer for the upkeep of the water system. Consequently, handpumps are often left unmanaged awaiting their inevitable breakdown (van den Broek and Brown, 2015, van den Broek, 2017; Brown and van den Broek, 2016). In our opinion, unless there is a resolution to establish a solid
Broek, 2017). Affordability of post-construction maintenance is an important consideration when planning the construction of sources.
13 management structure and a reliable system to guarantee the financing of handpump maintenance, the UN 6th Sustainable Development Goal effort to achieve universal and long term access to clean water remains, in our opinion, a mirage.
Figure 2: A broken and abandoned handpump in study area
3.3 What are the recurrent costs of handpumps?
Post-construction sustainability, in a context where the user is responsible for maintenance, necessitates careful planning at the source construction phase. With its mechanical parts, handpumps require regular maintenance and can expect repairs and possible rehabilitation over its lifetime. Recurrent costs can be divided into minor maintenance and major maintenance costs. A study by IRC, termed the WASHcost benchmarks, calculated the costs of maintenance for water services in developing countries. The estimated recurrent costs for water sources equipped with a handpump are shown in table 6 (Burr and Fonseca, 2013, p. 63).
The costs of repairs: labour and spare parts need to be paid for out of the user fees. Thus, there needs to be enough paying users to cover the likely costs of a handpump repairs over its lifetime. The desire by government and donors to extend the supply network must not be at the expense of ensuring that there is a robust system in place for ensuring enough funds are likely to be available long term. This is a very important consideration in developing post-construction management systems.
14
Table 6 Estimated maintenance Costs of Handpumps
Cost components of maintenance Estimated costs per deep BH per annum • Minor maintenance and repairs 171,000 UGX (USD 60) Expenditure on labour and materials needed for routine maintenance • Major maintenance 342,000 UGX (USD 120) Renewal, replacement and rehabilitation costs • Total estimated costs per annum 513,000 UGX (USD 180) *USD 1 = 2850 UGX
3.4 Rural Water Management Models
Because local communities control the management of communal water sources, there is not one universal model that can generate sustainable rural water management practices, as every community and location is unique with different local characteristics and conditions. Hence, determining the ‘right’ rural water management model is a matter of local negotiation and experimentation.
To explore the effectiveness of different rural water management models in the Gomba context for enhancing reliable access to water, the research team examined five management options commonly practised or piloted in Uganda:
(1) Community Based Management;
(2) Village Savings and Loan Associations;
(3) CBM-lite;
(4) Mobile Banking;
(5) WaterTime.
The first four management models represent a community based management model, wherein communities both pay (after the collection of water) and control the management of the water source. The last model, WaterTime, redirects the control over the water source to a private operator and involves a pre-payment and handpump
15 monitoring system. The pre-payment and monitoring technology has been developed by the social enterprise Susteq in collaboration with the PRACTICA Foundation. WaterTime has proof of concept having been successfully tested on two handpumps in Uganda with two local partner organisations of PRACTICA Foundation (JESE and HEWASA).
The five management models are set out in the Appendix 1 of this document, and are illustrated in Figure 3.
Figure 3: From the upper left corner to the right: (1) Community Based Management, (2) Village Savings and Loan Associations, (3) CBM-lite, (4) Mobile Banking and, (5) WaterTime
16
4. Study Methods
This section provides an overview of the data collected for this study. The fieldwork took Maddu and Kabulasoke sub-counties in 12 randomly selected villages (by the District Water Office) (6 villages apiece) – with the proviso that we visited villages without safe-water coverage and those that either have or had a protected source, over the course of two weeks in February 2017. The case study villages are identified on Figure 4.
In each village, a community meeting was organised by the Gomba District Water Office with a total of 375 participants. The researchers were accompanied by a representative of the District Water Office, as well as on a number of occasions the Assistant District Water Officer. A Water Compass representative was present at all meetings and acted as our main translator.
The aim of the village meetings was to assess current water arrangements and management practices and to examine, in a participatory fashion, how handpumps could be sustainably managed in Gomba District. At the outset of the community meetings, the researchers carefully explained they were from the University of Portsmouth, that they had been asked to conduct a study for the Water Compass and had come to learn about current water availability and management practices (if a handpump was present) and in particular to provide recommendations on possible management models.
17
Figure 4: Map of Study Area
18
Figure 5: van den Broek explaining the purpose of the research, with Brown and a representative of the District Water Office holding the poster. The Water Compass Engineer and translator is seated on the far left.
We stressed that our presence in their community did not signal that the Water Compass would be prioritising them for a new water source. Whilst we had asked for randomly selected villages, we did find we were taken to villages that did have very poor to non-existent safe water coverage. In Maddu sub-county on two days the political representative – the LC3 – did accompany us and made statements that we were Water Compass, who was going to bring water to their community and it was down to his work that we were there. We then had to once again stress our role in order to better manage and not raise expectations.
19
Figure 6: Community members in a participatory meeting
Figures 7 & 8: Recording data from participatory meetings.
Permission was asked to take photographs and to use a ‘go-pro’ camera to discreetly film proceedings for later consultation.
20
The fieldwork took place towards the end of the dry season – February 2017 – when observable levels in the valley tanks and dams were very low. It needs to be highlighted that during a water stressed period, tensions and concern over water availability is likely to be higher than if the study had been conducted during the wet/rainy season.
All community meetings followed a similar format – after introductions we asked contextual questions and then discussed their current water arrangements – where they sourced water from, if they paid for water vendors and how many jerry cans they bought. If there was a constructed water source, we asked about the management system in place and any experiences of handpump breakdown. We then asked community members to make suggestions over how they thought a water source could be best managed into the future.
Our experiences in other districts has been that female participants in community meetings tended to be very reticent and often separate meetings had to be arranged to capture the female voice. In Gomba, the pattern in many villages was for the men and women to sit separately, with the former on benches – in Figures 6,7 and 8 one can see men in the frame on wooden benches; and in Figure 9, women participants seated on coverings on the ground.
We then introduced the five different management models – see Figure 10, and to better explain different elements of the models role-play was often utilised – see Figure 11. There was the opportunity for questions and clarifications. We then asked the gathered participants to consider which model they thought would be most appropriate and to vote for it – we stressed this should be an individual decision (see Figures 14 and 15). We noted the number of votes for each model (and filmed the voting process) and then asked for representatives to explain why they had voted for a particular model and what modifications, if any, they would recommend.
21
Figure 9: van den Broek explaining the different management models
Figure 10: Demonstrating the WaterTime model – Brown represents a handpump!
22
5. Findings
This section documents the main findings of our study. We first provide our assessment, based our two week observation period, of water availability in Maddu and Kabulasoke sub-counties, making links back to the District Water Office reports on section 2.2. We then set out the key findings of our participatory meetings discussing management arrangements for the sustainability of constructed water sources.
5.1 Assessment of current water arrangements and management practices in Maddu and Kabulasoka
As discussed, we undertook fieldwork towards the end of the dry season when water levels in the valley tanks and ponds were low – refer to Figure 1 – which needs to be acknowledged when making our assessment.
Figure 11 Women collecting water from a pond in Maddu sub-county
23
As outlined in section 2.2, the operation area of The Water Compass in Gomba District (Kabulasoka and Maddu sub-county) is largely arid and marked with a low water table. As Table 2 indicates, the two sub-counties have low safe water coverage (40.5% in Kabulasoke and 50% in Maddu). In the locations we visited, protected water is mainly accessed from deep boreholes equipped with a handpump. The area has multiple unprotected water sources such as valley dams and open ponds (refer back to Figures 1 and 11). Figure 12 illustrates the high sediment load of the water during the dry season.
“Only when it rains do we get clear water” (Rwakibira, Maddu Sub-county)
Particularly in the dry season (December-February and June-August), when most open water sources run dry, rural communities reported they struggle with the collection of sufficient water for drinking and household use. Further:
“People get water from the pond – they drink with their animals. People and cows – they share” (Rwakibira, Maddu Sub-county)
The sharing of open water sources with livestock and wild animals was reported in several villages, including Rwakibira and Butambazi.
24
Figure 12: Mugya community member showing quality of water from village dam
In this period, water is treated as a commodity that needs to be paid for. We found that most households in our study area buy water from water vendors (see Figure 13) that collect water from contaminated and/or protected water sources7 at a price ranging from 300 shillings to 1,000 shillings per jerrycan (20 litres). Community members in the 12 studied villages reported a household expenditure on water between 5,000 to 10,000 shillings per week during the dry season. The cost of water was also resulting in some households limiting their consumption. Those households unable to afford the water vendor charges reported they collected water from distant sources: the implication for vulnerable and elderly households is a concern.
7 Community members purchasing water from vendors did so on trust – they did not know the origin of the water – meaning it may not be from a protected source.
25
Of the 12 studied villages, 4 villages had experienced access to safe water from a borehole equipped with a handpump – however, at the time of our study only two sources were functional – see Table 7.
Table 7: Overview of study villages experiencing access to safe water
Village Borehole? Additional Information 1. Mugya (Maddu) No 2. Rwakibira No 3. Nakasetta No 4. Bujunjuzi No 5. Kyabagamba Yes Most recently out of operation for 4 month (October 2016). 6. Gwanika No 7. Kakipuuya (Kabolasoka) Borehole Additional borehole half a mile away out of according to community members. operation for 6 years 8. Kisamula West No 9. Kawoko Lusosi No 10. Butambazi No 11. Mityegomga B Yes Source has been repaired – village borrowed money and are yet to repay. 12. Lwebajjo Used to Do have access to a borehole in next village, have a about a mile along according to community functional members. borehole
With the exception of one handpump in Kakipuuya, community members did not pay for the operation and maintenance of the service and Water User Committees (WUC) struggled to fulfil their responsibility of collecting funds and maintaining the source.
“People don’t pay. Therefore we collect money when it is broken. The longest breakdown period was four months when the pipes and cylinder got broken. It was a lot of money we were required to pay, about 200,000 shillings for the repair. Because the other boreholes are far, people used the water from the swamp.”(Kyabagamba)
The resistance of community members to pay for a seemingly functional water source was reported in all villages that either had a functional source or did have a source in the past. The result of waiting until a source is broken can be protracted handpump downtime whist the arduous task of “mobilising people to collect money” (Lwebajjo) is undertaken, meaning
26 without an alternative safe source, community members revert to accessing untreated surface water. In Mityegomga B, the Water User Committee Treasurer commented “getting funds takes longer than waiting for the Hand Pump Mechanic” – he also reported “people refusing to pay threatened to take me to prison!”
Figure 13 A Water Vendor on a motor bike
An important part of the study was to gauge understanding of the costs associated with maintaining a deep borehole, acknowledging that any management system and user fee has to be realistic depending on the depth and number of pipes in the borehole construction. All participants in the 12 study villages understood the need to pay for the operation and maintenance of the water infrastructure as the following quote illustrates.
“A borehole is like a bicycle – things need repairing and greasing – this is why we need to pay – in case it is broken it can be repaired straight away.”(Gwanika)
27
We gathered community views on how best to manage constructed water sources for long- term sustainability. With the acceptance that users need to pay for maintenance, a range of suggestions was made by both male and female community members: these ranged from communal management (to be expected given it is widely known, even in villages without a source) to a tendering system (a community member is awarded the right to manage a public good such as a market). The former suggestion was frequently proposed by elderly members of the community, with the latter suggestion often coming from the young men in the meetings, though not exclusively. In Nakasetta one proposal was for all users to pay a 200 shillings a week ‘membership’ fee: this would entitle them to one jerry can a day – additional jerry cans were to be charged at 100 shillings. Paying a monthly amount was often proposed in many villages, with a 1,000 shillings tariff frequently suggested. Some villages proposed paying per jerry can rather than a flat fee (100 to 200 shillings was regularly mentioned) – this is because “some are in business” (for example brick making which utilises a lot of water) and would benefit unfairly from a monthly fee Interestingly, in all villages the proposed water user fee was considerably less than the amount users were currently paying for water vendors, and also would not be sufficient to cover O&M costs associated with a deep borehole. For example in Gwanika village a community member proposed a monthly water user fee of 1,000 shillings per household – he then admitted he was currently spending 10,000 shillings a week during the dry season for water that may or may not be from a protected source.
Gomba’s relative water stressed situation, compared with other areas we have studied in Uganda (where there is a higher coverage and availability of protected springs) together with experience of paying water vendors, means that a management system that does enforce the user pays principle may have a level of acceptance. However, the experiences of those villages with boreholes indicates issues with community based management and a mindset of only paying once a source is broken needs to be considered. Ensuring vulnerable groups have access to water is also a consideration for any management model.
5.2 The Selection of The Water Management Model After a thorough situational analysis of the village and a brainstorming session on how to organise a reliable management structure for communal water sources, community members were introduced to the five pre-selected rural water management models by the research team (see Figures 3 and 9). In order to better illustrate important elements of the
28 different models, role-play was utilised – see Figure 10. Utilising familiarity with purchasing mobile phone credit – known locally as ‘air time’ – was a very useful parallel when explaining the fifth model, which we began to call WaterTime.
We encouraged participants to ask questions about the models and to clarify any areas. Participants were really engaged and asked numerous pertinent and perceptive questions that indicated that they were really considering the implications of the models for their communities.
A number of excellent issues were raised including the realisation that the voluntary nature of the Community Based Management model was potentially problematic:
“They can’t work for free – they all also need something” (Kawoko Lusosi).
In Kakipuuya a community member commented: “who thinks it is easy to collect money? That is not easy – it takes time”.
In Kawoko Lusosi the security of the collected fees was raised: “Who will keep that box?”
Other participants in Kakipuuya suggested with regards to introducing CBM: “that will bring fighting and violence”. Another stated that CBM “gives them headache and is disturbing them”.
Turning to the VSLA model, a commonly raised question was “what happens if they borrow all the money and the borehole breaks down?” (Kawoko Lusosi). We explained that was a risk with the VSLA model and that it was vital that a reserve amount was kept and not to have loaned out all the money. In Kakipuuya concerns were raised that funds may not be repaid: “People are tired of thefts”. A participant in Kisamula West rasied a similar concern: “If there is no bye law and they do no repay the loan what do we do?” It was agreed that: “a verbal bye law it is not good enough: it need to be written and given to the sub-county” (technical official). While participants recognised the system with keys was for security they thought it was “too much bureaucracy” (Kakipuuya).
The CBM-lite model raised some comments. In Bujunjuzi we were asked: “how do you know the money collected by the caretaker is all the money he has collected?” We were asked the cost of opening the SACCO account. In Kakipuuya we were asked to clarify the process of selecting the Water Operator: “these people are being paid – who allocated?” We explained that in this model, the water operator is selected in a village meeting from a short list. The Kakipuuya meeting also raised issues about the clustering of sources and cross subsidisation: they wanted to clarify that “their money is used to repair another
29 source? How do they benefit?” We explained that if their source broke then it would be repaired with pooled funds from the other sources managed by the water operator in a quid pro quo arrangement.
In terms of the fourth model, Mobile Money, clarification was sought in several villages that the District could not access the collected funds. We explained that the role of the District was just to provide some oversight, and that they did not have access to the funds on the mobile bank account. This did not seem to really satisfy people. Other questions were asked over “who monitors the caretaker and how are collected funds accessed?” (Kakipuuya)
Model five – WaterTime received the highest level of interest and great number of questions: after outlining the model, in most villages, participants clapped! Commonly raised questions included whether the electronic tags were provided free of charge and “what happens if you lose the card”? (Mugya). Here we again found the parallel with mobile phone useful – participants understood that if you lose your airtime scratch card or phone loaded with airtime you lost the credit and phone. The importance of looking after the electronic tag as per one’s mobile phone was recognised. In Gwanika we were asked “how many Hand Pump Mechanics can repair the technology? What happens if he may be one?” Ensuring the HPMs are trained is a very important point. In several villages, concern about understanding how to operate the new technology was raised by the more elderly attendees. Overall, the new technology did not appear to perturb participants: “same as air time they can learn the system” (Bujunjuzi). However, given that accessing water is so vital it is imperative that training is provided. Our later visit to the test sites in Fort Portal reassured us that the system is simple to use and no reports of difficulties were made. Again, the more elderly participants raised questions over their ability to pay for water: in Gwanika a participant said “some people can’t afford water time – will they just starve and die?” We explained with the WaterTime technology it was possible to make provision for elderly and disabled community members – a form of subsidy could be developed. The availability and location of the kiosks providing the service of topping up the electronic tags with WaterTime credit was raised in many of the villages. Finally, the security of the solar panels on the WaterTime device was raised in the meetings – some participants had experienced having their personal solar panels stolen. The security of the panels is an important consideration. In the test pilots, the panels are removed in the evening.
After we had clarified and answered questions on the five models, communities were invited to vote for the most feasible management model in their village context. Participants
30 were asked to make an individual choice. Table 8 provides for an overview of the outcomes of the voting in our 12 study villages. Of the 375 research participants, 319 people voted for the WaterTime model.
Table 8: Voting patterns by village for the five management models
45
40
35
30
25
20
15
10
5
0
CBM VSLA CBM-lite Mobile Banking WaterTime
Figures 14 and 15 People vote for their preferred water management model - WaterTime
31
Figures 14 and 15 show the popularity of the WaterTime model in two different villages. In three villages, community members voted unanimously for the WaterTime concept. In the other remaining nine villages, the majority voted for WaterTime. Participants were then asked to explain their choice. Reasons for introducing the WaterTime model are now listed which we have grouped into categories.
WaterTime will not cause internal conflicts “We have been quarrelling with people and have been taking them to the sub-county. Now, people will quarrel with handpumps.”( Mityegonga B)
“There is no relationship in this model; that I can decide to give you water or not. This system is independent, nobody can access the water without paying.” (Kawoko Lusozi)
“I used to be the water user committee chairman of the handpump in the neighbouring village. Kids and adults used to abuse us and undermine us. It is now the machine people can abuse.” (Lwebajjo)
“It does not give headaches and wrangles with people – you can pay as much as you use” (Kisamula West).
32
WaterTime ensures payment for operation and maintenance
“There are some people who don’t want to pay, in this model there is no favour.” (LCI, Gwanika). In the other models through personal and social relations it is possible to bypass the system of payment and still access water.
WaterTime enables timely repairs
“This model does not give us a hard time. The mechanic will come and there is no need to arrange transport and a HPM. Everything is automatic.” (Kisamula West)
WaterTime removes the need for voluntary inputs
“People don’t come to meetings nowadays. People have so many commitments, they don’t have time.” (Kyabagamba)
WaterTime brings development in the community
“When the phone method came, people used not to come close to those who had phones. The world is changing and also with this method people will cooperate.”(Lwebajjo)
WaterTime was supported because it is seen as “modern” – the desire for new technology was clear; it was also the reason why the Maddu LC3 supported it.
In addition to the village meetings we also interviewed a representative from the Gomba Distrcit Hand Pump Mechanics Association who accompanied us on the research, representing the District Water Office, as well as the Gomba District Water Officer for his views on the research and different management models. The handpump mechanic’s view is that “few villages are managing their sources well…. The Water User Committees are not active even though we teach them - they don’t perform well”. He reported that he often struggles to be paid for repairing boreholes: for “rehabilitation they fail to pay – they eat money, the Treasurer and even the WUC eats the people’s money…..you fix it and they get water and they say they will give you tomorrow” – some villages have delayed payment three or four times. He was strongly in favour of the WaterTime model because he foresaw that funds would be available to pay for repairs: “handpump mechanics – we are going to get jobs – we know very well a salary will be coming to us. You can spend one month without a call to repair because they don’t have money”. He suggested that if the model
33 was adopted, training should be provided to the hand pump mechanics in how to repair the innovative technology.
The District Water Officer’s experiences with the Community Based Management system were not favourable: “every quarterly report I write these WUCs don’t work. All the time the stress comes back here”. After presenting the five models to him his view was that the WaterTime model had potential that was worth investigating – the other four models were in his view flawed because of the issue over collecting and storing funds.
5.3 Visiting Demonstration Sites
The WaterTime model received a lot of support in Gomba, so visiting the NGOs (JESE and HEWASA) piloting the technology in the Fort Portal area was important in order to learn more about the system - known locally as “pay as you fetch” - and how it might operate in Gomba. A joint meeting was held with JESE and HEWASA officials, before we visited the pilots, to understand how the pilots had started. The selected sites saw their boreholes, one of which had been down for between 4-6 the other 1.5 years, rehabilitated. Practica engineers then installed the technology on two selected sources in October 2016. The system, which also includes the running of the cloud dash-board, can be seen in
Figures 16 and 17.
Figure 16 The WaterTime Technology at the pilot site in southwest Uganda
34
Figure 17: Kiosk owner loading WaterTime credit to Susteq/Practica designed technology – the tag is returned to the owner who is able to collect water from the handpump as it can read there is enough credit.
Prior to commencing the pilot, community development officers from JESE undertook surveys to determine how much people were paying for water since their borehole had broken down. They found households were paying 500 shillings a jerry can from another source, and used on average 4 cans a day. When the handpump was operational it was reported by members of the Water User Committee at a meeting organised by JESE that illicitly collecting water at night, without paying for it, was commonplace: “some could refuse to pay – they say they had not used the borehole and yet they collect at night – now that can’t happen – they need the token”. As with the village meetings in Gomba “getting money from people is not easy”. In the JESE pilot village every member of the community is part of the pilot. Each household (104) was given an electronic tag – replacements were to be charged at 10,000 shillings – in the 4 months the pilot had been running only one person had reported that their tag had been stolen. The pilot was charging 100 shillings for 20 litres8 of water –– this was considerably cheaper than the 500 shillings a jerry can they had been paying, with the kiosk owner suggesting “this was a great favour to them”.
8 Note – jerry cans are about 22-23 liters.
35
A water committee member suggested: “there is no-one who can’t afford”. The smart technology on the pump means that once 20 litres is reached the handle of the pump automatically locks until it is tagged again and credit removed.
A hybrid management system had developed - a water committee oversees the activities of the kiosk owner who is entitled to keep 10% of the collected fees. The collected money then goes to the water committee - they opened a bank account were collected funds are stored. When we questioned how transparent and safe the funds where – in essence the whole crux of the management system is ensuring funds are secure and available for repairs – JESE and HEWASA officials strongly feel: “it is not privatised it belongs to the committee”.
The location of the pump close to the kiosk meant the kiosk owner could monitor activities and was able to provide assistance to those who need it. Further, if the kiosk owner has to travel, another shopkeeper across the road had been trained to operate the system so people can always load credit. The kiosk owner suggested everyone was still participating in the pilot: “there is no one who has left the source – at first some were against it”. National Water were constructing a piped water system in the pilot village but we were told the advantage of the pilot was it was “consistent – always on 24 7 unless there were mechanical problems”, whereas National Water had a reputation for intermittent supplies and also a high connection fee.
Overall, community representatives suggested the pilot has improved relations and the assembled committee members suggested they had “not met any challenge and were happy with it”, the system was “very easy” to manage and to collect water, and they would “very fine truly recommend it – any community can be grateful – you can come at night and collect if you have money on your account”. The kiosk owner had developed a timetable for when he needed to be in the kiosk to top up customer tags with credit, and to receive the credits from the network. The optimum time to send and receive data is an important aspect for consideration. The cost of the monthly mobile bundle is 5,000 shillings –the kiosk system communicates with the system dashboard four times a day. The cloud ‘dash board’ shows usage of the handpump and functionality, and logs how much credit has been bought, which can be cross-referenced with usage at the pump. At present only the NGOs and Practica have access to the dash-board.
According to both HEWASA and JESE the biggest issue they had experienced with the pilot had been the supply chain – one pump was out of operation for two weeks whilst waiting for spare parts from the Netherlands to repair the technology.
36
Both NGOs reported that local government and local leaders were supportive of the pilot, because they no longer have to call meetings about non-payment: it has “saved a lot of time and people don’t fight”. Ultimately “they are paying for a service and a service is there”.
6. Conclusions and Recommendations
The pioneering study highlights that community members in the 12 study villages in Gomba District rejected the dominant community management model in favour of a model (WaterTime) that enforces pre-payment to cover operation and maintenance costs of handpumps, at the same time removing voluntary community control over the management of the handpump. The study revealed enormous support amongst community members for introducing the WaterTime concept in Gomba District. A local village leader in one of the villages said: “We kindly request the district and the Water Compass to get us the pre-payment model, it will stop the conflicts in this village” (local village leader Kakipuuya). The model also has the support of the District Water Officer and Hand Pump Mechanic Association in Gomba.
The JESE and HEWASA pilots have demonstrated proof of concept in terms of the technology. While “pay while you fetch” does indicate its pre-paid nature, it does not in our view do justice to the innovative technology. We also found using familiarity with purchasing mobile phone credit – air time - really useful when explaining the model – which is how the term WaterTime developed. We feel there is considerable potential to develop the system further with an enhanced management system. We feel that having a watertight system for collected funds is vital or trust in the system may be eroded. The hybrid system in operation at the demonstrations sites, where the water committee was in charge of the collected funds meant there was, in our opinion, a potential lack of transparency. We would recommend that funds are sent by mobile money to a bank account that in the first instance is managed by the Water Compass. The amount of remuneration for the kiosk owner also requires some consideration – the 10% of collected fees seemed rather arbitrary. Having the kiosk where WaterTime credit can be loaded near to the source was beneficial at the demonstration sites - they were able to monitor activities and help young children and the elderly get used to the system. We would recommend careful consideration is given to the
37 siting of the kiosks providing this service. It is also vital that if there is a safety net in case the kiosk owner has to travel – in the pilots having a second kiosk owner trained in using the system meant people would always be able to top up their tags. To help reduce accidental loss it may be worth investing in lanyards so that the tags can be worn around necks. It is crucial, as JESE and HEWASA suggest, that practical ways to replicate spare parts in Uganda is explored, as well as training in maintenance of the system. While the technology was developed to be retrofitted to hand pumps, this study recommends the Water Compass to install the WaterTime technology from the outset to avoid having to accept a new management system.
Gomba district has a number of favourable conditions for piloting the WaterTime concept:
1) The arid geography - the area has limited water alternatives which could undermine the functioning of the model; 2) The concept of paying for water and paying a private individual for water delivery is already accepted. With WaterTime, payments would be considerably less and water would be collected from a protected water source. 3) There is a wide consortium of support for introducing the WaterTime model from water users, Handpump Mechanics and District officials. This high level of support was not expected at the outset, but once again underlines that there is a demand for a new model that depersonalises payment.
As a result we feel that it is worth considering investing in some WaterTime units and carefully piloting them. The pre-paid mobile-enabled technological innovation and management system that uses familiarity with purchasing mobile phone credit, has the potential to transform the failing and underserved rural water system in Uganda into a viable business, ensuring post-construction maintenance of boreholes for sustainable and equitable access to water. WaterTime has the potential to sustain initial investments. The average costs of drilling a borehole is USD 12,000 and the projected cost of the WaterTime hardware is USD 1,200 - representing 10% of the initial investment costs. As discussed above, introducing the WaterTime concept also entails additional training on installation and Handpump Mechanic training as well as ensuring a watertight system for managing the collected funds.
To date, donors have spent billions of dollars on extending the supply network. There is an urgent need to focus on the sustainability of rural water sources and on the basis of our research, the WaterTime model offers considerable potential to ensure initial hardware investments are sustained. This is an investment that may support communities in gaining
38 reliable access to water and that may further the ambitious Sustainable Development Goal of achieving universal and sustainable access to water.
References
Adams, W.M, (1992); Wasting the Rain. Rivers, People and Planning in Africa, Earthscan Publications Ltd, London UK. Baumann, E., 2006. Do Operation and Maintenance Pay? Waterlines 25 (1), 10–12.
39
Fonseca, C., 2014. The Death of the Communal Handpump? Rural water and sanitation household costs in lower-income countries. PhD Thesis. Cranfield: Cranfield University. Jones, S., 2011. Participation as citizenship or payment? A case study of rural drinking water governance in Mali. Water Alternatives 4 (1), 54-71.
Rural Water Supply Network (RWSN), 2010. Myths of the rural water supply sector. Perspectives Paper No. 4.
40
Appendix 1 Water Management Models
Management Model Community Village Savings and CBM-lite Mobile Banking WaterTime Based Loan Associations
Management Resource management organisation Organisational Elected An elected WUC and paid Water Operator is WUC with a paid WaterTime is social business structure voluntary Water caretaker. Community responsible for a caretaker. model and a pre-payment and User Committee members (ranging from 15-60 cluster of water monitoring technology for (WUC) (6-10 people) form a savings group points with a handpumps. The system members) for (which includes the WUC). The caretaker at each operates without a WUC/water each water group set a monthly interest point (salaried) operator at the pump. Water point in a rate (5-10%) for loans taken users pay per volume of water village. out. At the end of the year, pumped. GSM connectivity of members receive a return on the unit stores data on an their savings ranging from 30 online database to send alerts to 60 percent annually in case of malfunctioning of generated from interest and the pump. A water credit fees collected throughout the vendor sells credits and uses year. After 9-12 months an a solar powered ‘shop unit’ to ‘action-audit’ takes place and charge water credits on a the money is shared among token. the members. Members can only join if they Personalized tokens and the also pay for the O&M of the software allow for tuning the water source. pre-payment unit to local needs, such as fixed amounts of free water per day and flexible tariffs (per person, per time of the day, per season, block tariffs).
A ‘Micro Water Service Provider’ (MWSP) is a central player in this model. The for-
41
profit MWSP is responsible for the operation and maintenance of the rural water infrastructure. The revenue of the water credits cover maintenance, repairs and depreciation of the infrastructure. Resource Governance
Financial All households Members select three Periodic fixed Households pay to It removes the need of Accountability (registered individual who are entrusted payments to a micro- water operator handling cash at the water exemptions) with a key to one of the three credit organisation, (possibility of mobile point. Funds are directly expected to pay locks on the cashbox where payment slips and banking) who wires transferred to the MWSP. water source the group’s funds are kept. All access to the the money with caretaker 1,000 transactions –the collection of account is granted mobile banking to a Ugandan member savings and the through approval of bank account Shillings (UGX) disbursement of loans – are various signature (centenary bank). per month. Fees carried out at weekly meetings holders. The account is stored by the in front of all members, supervised by the WUC within ensuring transparency and district and members community. accountability. cannot access the funds without signatories.
Key incentives Non-monetary Membership VSLA and ability Water Operator Paid caretaker Financial incentive MWSP. rewards - WUC to take out loan and applies for the Communities have to first pay members acting accumulate savings (financial position (through a (an affordable and small altruistically incentive members). The letter to the local amount) to access the water. caretaker is paid from the O&M government) and savings. receives, with the caretakers, a financial compensation from the water user fees. Key sanctions Social pressure Social pressure and removal Social pressure and Social pressure and Without payment no access to and community from VSLA. Creation of by- communally agreed creation of by-laws. the water. formulated laws. by-laws enforced by
42 graduated Water Operator with sanctions to local village leader. curb free-riding Contract stipulate culminating in responsibilities and exclusion from sanctions signed by source. key actors. Recourse to Sub-County officials (higher level)
43